Selected work

Representative Papers

Does More Bandwidth Really Not Matter (Much)?

Mar 05, 2025Proceedings of the 13th EAI International Conference on Mobile Multimedia Communications, Mobimedia 2020, 27-28 August 2020, Cyberspace

This paper challenges the conventional wisdom that bandwidth is negligible and latency dominates web performance, systematically investigating the actual impact of mobile network bandwidth and latency on modern complex web pages. Method: We propose a novel metric—Critical Path of Improvement (CPI)—to quantify websites’ sensitivity distributions to bandwidth versus latency. Leveraging real-world network measurements across four operators and 57 cities, we construct an empirically grounded CPI model and perform critical-path decomposition and correlation analysis. Contribution/Results: Our findings reveal that 18% of websites are predominantly bandwidth-limited, while over half exhibit significant bandwidth sensitivity. The study uncovers the underappreciated role of bandwidth in mobile web performance and delivers a practical, deployable framework for diagnosing bandwidth sensitivity—providing empirical guidance for developers optimizing resource loading strategies and operators designing differentiated QoS policies.

1 citationsRead paper

Detecting and Characterizing Massively Shared IP Addresses

Aug 06, 2026

This study addresses the significant disruption caused by large-scale IP address sharing—such as carrier-grade NATs and proxies—to IP-based attribution, blocking, and rate-limiting mechanisms. For the first time, it systematically identifies and characterizes such shared IPs at a global scale by leveraging diurnal traffic patterns from CDNs, behavioral clustering of IP addresses, and network-level features including cellular connectivity and dual-stack capability. The findings reveal that over 40% of IPv4 traffic originates from fewer than 2% of active IPs, with sharing particularly pronounced and growing in smaller countries. While IPv6 exhibits substantially less sharing overall, mobile networks present a notable exception. This work provides critical insights into the global distribution, origins, and evolutionary trends of IP sharing, offering foundational implications for network measurement and security policy design.

0 citationsRead paper

Recent publications

Latest Papers

Detecting and Characterizing Massively Shared IP Addresses

Aug 06, 2026

This study addresses the significant disruption caused by large-scale IP address sharing—such as carrier-grade NATs and proxies—to IP-based attribution, blocking, and rate-limiting mechanisms. For the first time, it systematically identifies and characterizes such shared IPs at a global scale by leveraging diurnal traffic patterns from CDNs, behavioral clustering of IP addresses, and network-level features including cellular connectivity and dual-stack capability. The findings reveal that over 40% of IPv4 traffic originates from fewer than 2% of active IPs, with sharing particularly pronounced and growing in smaller countries. While IPv6 exhibits substantially less sharing overall, mobile networks present a notable exception. This work provides critical insights into the global distribution, origins, and evolutionary trends of IP sharing, offering foundational implications for network measurement and security policy design.

0 citationsRead paper

Does More Bandwidth Really Not Matter (Much)?

Mar 05, 2025Proceedings of the 13th EAI International Conference on Mobile Multimedia Communications, Mobimedia 2020, 27-28 August 2020, Cyberspace

This paper challenges the conventional wisdom that bandwidth is negligible and latency dominates web performance, systematically investigating the actual impact of mobile network bandwidth and latency on modern complex web pages. Method: We propose a novel metric—Critical Path of Improvement (CPI)—to quantify websites’ sensitivity distributions to bandwidth versus latency. Leveraging real-world network measurements across four operators and 57 cities, we construct an empirically grounded CPI model and perform critical-path decomposition and correlation analysis. Contribution/Results: Our findings reveal that 18% of websites are predominantly bandwidth-limited, while over half exhibit significant bandwidth sensitivity. The study uncovers the underappreciated role of bandwidth in mobile web performance and delivers a practical, deployable framework for diagnosing bandwidth sensitivity—providing empirical guidance for developers optimizing resource loading strategies and operators designing differentiated QoS policies.

1 citationsRead paper